Problem

Drug discovery teams have increasingly powerful biological foundation models, but the models alone do not produce decisions that scientists can trust, reproduce, validate, or defend. Helical targets the costly gap between computational prediction and biological action: conventional discovery and wet-lab testing can take months or years, delaying program selection, consuming scarce experimental resources, and slowing treatment development. Its clearest use case is safety and translational decision-making, demonstrated by Pfizer’s use of Helical’s Virtual Lab and foundation models to discover predictive blood-based biomarkers for gene-therapy safety and help identify the right patients for treatment sooner and more safely.

Product / Service

Helical provides a virtual AI lab and an application layer that orchestrates biological foundation models against a customer’s own biology and disease context. Its Virtual Lab is aimed at biologists and translational scientists, while its Model Factory serves machine-learning engineers and data scientists. The system personalizes models, runs scalable and reproducible in-silico experiments, validates outputs against biological evidence, and ranks the candidates most worth advancing into a program. Helical also maintains an open framework with a unified interface for models spanning genomics, transcriptomics, and single-cell data, making model access, fine-tuning, and downstream experimentation easier.

The delivery model appears to combine software with scientific deployment and collaboration: Helical’s engineers work alongside customer scientists, beginning with proofs of concept and expanding successful work into production programs. The intended benefit is to turn months of wet-lab testing into hours of AI-driven experimentation and, according to the company’s reported deployments, compress parts of discovery from years to weeks while retaining a therapeutically relevant evaluation that the customer owns.

Market

Helical operates in the emerging AI drug-discovery and computational-biology market, specifically at the application layer between biological foundation models and pharma workflows. Its closest competitive set is other virtual-lab and model-orchestration platforms that convert biological data and AI predictions into reproducible, decision-ready experiments. The evidence does not establish a definitive direct-competitor list; Inductive Bio is a named adjacent player developing virtual chemistry labs, while the models integrated into Helical’s open framework—such as Geneformer, scGPT, Tahoe-x1, and Evo 2—are better characterized as an ecosystem of models Helical can orchestrate than as direct competitors.

Helical is early but has meaningful commercial and ecosystem traction rather than looking simply pre-revenue. It announced a $10 million seed round led by redalpine in 2026, reports production deployments with multiple top global pharmaceutical companies including Pfizer, and has announced work with Tanabe Pharma America on in-silico perturbation experiments across ALS and related neurodegenerative diseases. Its open-source repository had 227 stars and 38 forks in the cited snapshot. The available materials do not disclose revenue, pricing, or contract economics, so the strongest evidence of traction is paid or production-oriented pharma deployment, strategic collaborations, funding, and open-source adoption rather than a reported revenue figure.